Ch2 Intro to Tensorflow/005 - Tensorflow Intro.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting MNIST_data/train-images-idx3-ubyte.gz\n",
      "Extracting MNIST_data/train-labels-idx1-ubyte.gz\n",
      "Extracting MNIST_data/t10k-images-idx3-ubyte.gz\n",
      "Extracting MNIST_data/t10k-labels-idx1-ubyte.gz\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[8]]\n"
     ]
    }
   ],
   "source": [
    "x = tf.constant([[2, 2]])\n",
    "y = tf.constant([[1], [3]])\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    result = sess.run(tf.matmul(x, y))\n",
    "    \n",
    "print(result)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "you can evaluate the result with `foo.eval()`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-2 -1]\n"
     ]
    }
   ],
   "source": [
    "x = tf.Variable([1, 2])\n",
    "a = tf.constant([3, 3])\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    x.initializer.run()\n",
    "    sub = tf.sub(x, a)\n",
    "    result = sub.eval()\n",
    "    \n",
    "print(result)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import math"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "1\n",
      "2\n",
      "3\n"
     ]
    }
   ],
   "source": [
    "state = tf.Variable(0, name='Ge')\n",
    "one = tf.constant(1)\n",
    "\n",
    "new_state = tf.add(state, one)\n",
    "update = tf.assign(state, new_state)\n",
    "init_op = tf.initialize_all_variables()\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    sess.run(init_op)\n",
    "    print(sess.run(state))\n",
    "    \n",
    "    for _ in range(3):\n",
    "        sess.run(update)\n",
    "        print(sess.run(state))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now implement something similar."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "60\n",
      "120\n",
      "480\n",
      "960\n",
      "3840\n",
      "7680\n",
      "30720\n",
      "61440\n",
      "245760\n",
      "491520\n"
     ]
    }
   ],
   "source": [
    "some_state = tf.Variable(15, name=\"awesome\")\n",
    "two = tf.constant(2)\n",
    "\n",
    "update = tf.assign(some_state, tf.mul(two, some_state))\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    some_state.initializer.run()\n",
    "    \n",
    "    for _ in range(5):\n",
    "        sess.run(update)\n",
    "        print(update.eval())\n",
    "        \n",
    "        # eval executes the update twice. Same with `run`.\n",
    "        print(update.eval())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[array([ 21.], dtype=float32), array([ 7.], dtype=float32)]\n"
     ]
    }
   ],
   "source": [
    "input1 = tf.constant([3.0])\n",
    "input2 = tf.constant([2.0])\n",
    "input3 = tf.constant([5.0])\n",
    "intermed = tf.add(input2, input3)\n",
    "mul = tf.mul(input1, intermed)\n",
    "\n",
    "with tf.Session() as sess:\n",
    "  result = sess.run([mul, intermed])\n",
    "  print(result)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# some explorations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYwAAAEKCAYAAAAB0GKPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzsfXm8XsP9//tzkwiRhFiySIRaErUVJdWNa19qp0XVUlXa\nWmr7Fj9ViVZbpdqi1FbVfutrTVElRImlVGLfSZSQSCLElkTkJnd+f5xn+swzzyyfmTPnWW7O+/W6\nr3uec2Y7c2bmPZ9lZkgIgRIlSpQoUcKHjmYXoESJEiVKtAdKwihRokSJEiyUhFGiRIkSJVgoCaNE\niRIlSrBQEkaJEiVKlGChJIwSJUqUKMFCSRg9HER0OBE9VHAelxHRmUXmoeRV+PukBhF9hYheShGW\niNYkoo+IiBhpbUtEbym/nyeibXil5oGIriGic5hh1yKibiIqx502RfnhegAqg8y/iOgDInqXiB4i\nos8rQQpdbCOE+L4Q4tzU6ToGmLZaPCSEeFgI8dmYsET0OhFtrzx/SwgxUPAXUP03nBBiYyHEg+yC\nFwNWuXWyK9Ea6N3sApTIByIaAODvAI4BcBOA5QB8FcCnzSxXIhCyAcY7m06SGVEvIcTSRuRVwgv5\n7Uu0EEoJo/0xCoAQQtwoMnwqhLhXCPG8EoaI6HwimkdErxHRrsqDYUR0GxG9R0SvEtFRlft9iWgh\nEa1S+X0mEXURUf/K73OI6MLK9X/VEnJmSEQnE9EcIppJREco+a1CRH8nog+J6DEi+qlDxfRA5f8H\nFTXMFxjvM5CIriKityvl+KlNfUNEZxPRTUT0FyL6AMDhRLQVET1CRO9Xyn4xEfWuhB9LRBdVrnsT\n0XwiOq/ye3ki+oSIVjbko6uGXieiU4jomUo+/0dEy+lhiejPAEYC+Hvl/U/VpS4iOoKIXqw8n0ZE\nR1vqskZaqeT7UeVvfiXNkZVnexDRU5UwDxPRJkoamxPRE5Xvdz2A5R35dRDRBUQ0l4imAfia9txY\ndiLqB+BOAGsQ0ceV50Nd36ZEgyCEKP/a+A/AAABzAfwJwK4AVtaeHw5gMYAjkc3avgdgpvL8QQAX\nA+gD4HMA3gHQWXk2CcC+leu7AUwFsEvl9wMA9qpcXwPgnMr1tgC6AJwNoBeA3QAsALBS5fn1AK4D\n0BfAZwG8CeBBy7utBWApAAp4n78BuBTZQLYagH8D+K4l/bORSWJ7Vn73BbA5gDGVtEcCeAHACZXn\n2wF4pnL9RQDTADxa+b09gKcs+WwL4E3l9+uVcg0BsDKAFwEc7Qi7naFOOiq/dwOwduX6q5W63syR\n1vaG8p0L4P7K99ocwBwAW1bq4NBKvD6VvzcAnFAJu3/lW5xjee/vVd5tjcp73hdb9sq9LWzfpvxr\nzF8pYbQ5hBAfA/gKgG4AVwB4pyIxrK4Ee0MI8UeR9bprAQwjosFENALZwHeaEKJLCPEMgKsAHFaJ\n9yCAbYmoF4BNAVxU+d0XwFYAbJLBYgA/FUIsFULcBWA+gNGVWfF+AH4iMknopUp5fNAlBNv7DEY2\nCJ0khFgkhHgXwG8BHOxI+1EhxN8BoFKmp4QQk0WGN5HV6bYyLID1iWgQgG0AXA1geGVGvA2qEhEH\nvxNCzBFCfIBMpbhZwPv/F0KIu4QQb1SuHwJwD7LBlwUiOhBZ/ewvMnXcdwH8QQjxeKUO/oKMVLeu\n/PUWQlxU+ba3AJjiSP7rAH4rhHi78p6/yFN2IcSTjm9TogEoCaMHQAjxihDiSCHESAAbI5vR/VYJ\nMlsJ+0nlsn8l3DwhxEIl7HQAwyvXDyCbVW8B4FkAEwF0Ihs4pgoh3rcU6T0hRLfye2Elv9WRzUxn\nKM9iDJu291kL2Sx4VkVd9T6APyCTNGyoyZ+I1q+ozGZV1FTnyvhCiEUAHkdWB9sgk8AeQUbY2yKM\nMOYo17J+gkFEuxHRo5SpFN9HRpiu91Xjbo5MutxHCDGvcnstAKdU6k/W4QhkbWUNADO1ZKY7slgD\ntfVbEza07K5vU6IxKAmjh0EI8Soy9dTGjOBvA1iFiFZU7o1EdVB4BMBoAPsCeEAI8XLl+e4IGxwl\n5gJYgmwAkljTET7U6PkWgEUAVhVCrCKEGCSEWFkIsWlAHpcBeAnAukKIlQGcidoZ/oPI1E+bIZtd\nPwhgF2QSVxEeSNY6qNg9bgbwKwCrCyEGAbgLDolEiTsYmfru+0KIZ5VHbwE4t1J/sg77CyFuADAL\n1cmExEhHNrNQ+33XCii76b1936ZEwSgJo81BRKMpMzAPr/xeE5mK4VFfXCHEDGSk8AvKjNybAvgO\ngL9Unn8C4AkAx6JKEI8g000HE0ZF6hgPYCwRrUBEG6Cq/jJhLjJV27rM9GcjU2v8hogGUIZ1KGzt\nwQAAHwkhFlbK933t+QOVMr8ohFiCTMo4CsDrQoj3AvLhYjaAdbR7cpBcrvL3rhCim4h2A7CzL8GK\nivFmAH+pqJVUXAnge0Q0phJ2RSLavTKpeBTAEiI6vmL03w+ZTcGGGwGcQETDK2q805RnvrLPAbAq\nEQ1U7vm+TYmCURJG++NjAF8A8BgRfYxsQH8WwKmOOOrs7WAAn0EmbdwC4CwhxP3K8weQqZEmK7/7\nI2w2reZ3PDID6Cxk9ofrYHEBrhDWuQD+VVGP2AYnNf3DkA1ELwKYh8zVeGhAWU8FcAgRfQTgcmRG\nehWPIDOoP1Ap44sAPkEYgYZITr8EcFbl/U9W4wsh5iMzQN9ERPMAHATgNka+IwB8GcCJFQ8k6Yk0\nQgjxBDI7xiWVNF9F5mgAIUQXMhvUtwG8h8xGoROOiiuROUs8g0yV99+wvrILIV4B8H8A/lN596Hw\nf5sSBYMyu2HORDK3xt8iI6CrhRDnac+/iers4mMAP5BisC9uiZ4NIvolgCFCiG83uywlSpRwI7eE\nUfF8uQSZHncjAAdXxEUV/wGwjRDicwB+hsy7gRu3RA9CRYW2SeV6DDIV2PjmlqpEiRIcpFBJjUHm\nMTO9IrJeD2BvNYAQ4t9CiA8rP/+NquHMG7dEj8MAAOOJaD4ylcP50q21RIkSrY0UqySHo9Z1bgbc\nhrCjkHlDxMQt0eYQQjwOYP1ml6NEiRLhaOiyeiLaDpnB7CuNzLdEiRIlSuRHCsKYiVpf7BGoX9yD\nisvmFQB2VRZ8seJW4pcbkZUoUaJEBIQQSdarpLBhTAGwHmWboi2HzD3udjUAZZua3QLgUCHEayFx\nVcTuf9LT/s4+++ym5PvGGwJA898/pi5++cvWK3sRdbHffgI339z8sjT7r5F9ZNEiga4uXlhAYNVV\niynHzJkCr75qyjMdchOGyPafOQ7ZgqkXAFwvhHiJiI6h6s6ZZwFYBcCllO2COdkVN2+ZShSDN99s\ndgni8ckn/jA9AePHA9dd1+xSLFtYYw3giCOaXQpgxx2BUaOKzSOJDUMIMQHZFhLqvcuV6+8iWwzE\niluihMQRRwDXXAP4z5dzI/FEq0SJ/2LePOCZZ/jhi2qL8+cXk66KcqV3G6Kzs7Mp+TZj0L32WmCp\n40gjbl0sC4Qh62JZeFcfmtVHejpKwmhDNLozdHUBixc3NEsA1YHPlXczCaO7G3jyyfTpxqIcJKto\ndF2EtK9GEfrSpcDbb6dNsySMEl7suSewqWu/14LAIYzQtFJiwgTg85/3hytRQkWjCOPdd4HNXKes\nRKA83rCEF5MnA+/bTr5oAFqVMD7tCaeml0iCvDa2lJg5E/joI2DQIKAjsUhQShglWhZFSBgPPdTz\ndfw9/f1aEXPnNrsE1e++887AhhtmKtPURFYSRomWRRGEsc02wPPP50+vRAkVc+YA011nDyrQCf2p\np9IO7IsWVfMpCaNEw9HsGetLCVbmqO/Q3W0PVyIMS5YATz/d7FK0BhYu9IcB6vvTyy+nL4vMp1RJ\nlUiGxYuBu+7yh5NoNHHI/PbYI11a+nWJfPjLX4DNN292KdobffoUk26pkiqRFOPHA7vv3uxS2JFy\nYG8WSXR3A5MmNTbPRr6rVH+U4EP/PqkJQ5JEKWGUSIplaaatvmsjPVomTwa2265x+aXEppsCxx7b\n7FIUjyVLgI8/zp9ObLsqJYwGYPvtgQMPbHYp2hut5ApoQlGE1kiidK1Sb3U89xxw333NLkXxGDsW\nGDiwefmnIgy9XZcShoL7788WTpUoHjEDrBDAHXc0Pt9GpLUsodUnFSnwn/80Nr+iVVISpYShoRwE\n8qHIweC117IV4nnQ6oSxLAym5TumR2nDKFFCQ6uReRHlabV3tGHaNODqq5tditZFs0mxN3O/jTff\nBFZd1R9OtstSwiiRFEV2lBSDqUwjxQysWYN7swcjAPjFL4Cjjmp2KVoXqb4RNx29Lfbqlf332bte\nfDHbSp2bbilhaGiXGV67o1n1LPPdccd0aaVEK5CBCSnf1feOrVoHIWj2O8jv1dXFC8dFKWFoaMRA\n9v3vF3+KVbMQOyNScfTRwKuvhsUJRUppJVV6JXoOUg2q3HZlC+fbAoebvmrDKAmjgejqAu6+G5g6\nNSzelCnABx8UU6ZWw5VXAn/7m/357NnxezfJDpKaMHo6SgmjFl1d2e6tNjTb6F2UhFGqpBqMb30L\neP318HhjxgCnn56+PK0KU4eTjfuAA4BNNolLtyjCaOQA0QoDap4ytEL5bfjb34BzzvGHO/FEYKWV\n7M9b5R19EoZvDzS9n7SsSoqIdiWil4noVSI6zfB8NBE9QkSLiOhk7dkbRPQMET1FRJND8i161vjs\ns/FxlyxJV45WQUx9pzgzolRJlTBh3Djg7LP94Rq1ziJ2cOZKGKHlKELCyH2AEhF1ALgEwA4A3gYw\nhYhuE0KoezC+B+B4APsYkugG0CmEaOIRPWa0ysyjKBT5finSLlVSaZDn3Vu5D3B3HfYNmq2ikkpl\nw5BoVQljDICpQojpQoguANcD2FsNIIR4VwjxBADTvJsSlcOL997L9q3noqeK8qGIXemdKt+SMJoH\nUzveemvgiCMaXpQ6cL9pqxOGhE8rwSVItd+0og1jOIC3lN8zKve4EAAmEtEUIvpuSMahg8AXvwis\nv35YnJ6MdiE1znd+553iyxGDVqjj1GV47DHgnnvSphmDdpUwurpqyYE7MQp9XoSE0Qpnen9ZCDGL\niFZHRhwvCSEeNgUcO3as8quz8mdGV1e2rfROO1Xvvf02sGABv2Ct0NnbAS6jdx6oK1Z9GDIkkyBX\nWcWdVqvg+OOzv1Z32W7lPhDqZhr7PDWEyLbN0c+iyUsYEp98MgnAJFx+OTBrVkwJ7UhBGDMBjFR+\nj6jcY0EIMavyfy4R/Q2ZistLGOPGudO9805gn32ap79t5Y4mYSrjnDnAkUcC//hH48ujI1Ql5TIa\ntprR+5JLgKFDgTPPbHZJ2hfNkDB++9vM62+HHfKlM2VK9Zo7MeK22379OgF04qijgGeeAWbN8gyW\nAUihkpoCYD0iWouIlgNwEIDbHeH/W61E1I+I+leuVwSwMwC2176rAps9KLQDYZjw2GMZ2apolg0j\nZVrNbg8mFNVG9Hd15fPyy9n23ja0cjtuBmGcdBLw4x/zw9uglilWJfX22/4xsOWM3kKIpQCOA3AP\ngBcAXC+EeImIjiGiowGAiIYQ0VsATgJwJhG9WSGKIQAeJqKnAPwbwN+FEGztaNGDQCt3lhQIfb9m\nHdHaU9dhNKocrvq7+mq3tN7KfYDbLuReTTY0ox+kUOMOHw7ccEN2vWRJvZdVS7rVAoAQYgKA0dq9\ny5XrOQDWNESdD2CzFGVQ8eijwCOP1N8P/SChDWmTTdyztZSYMQNYcUVg0KC06cZ0BlM9tbJbbSri\ny/uOjZIwXPANpr48mkkorWb0Dql3Nc88Kim5GeEBB1SvZdqt6lbbcvjyl4Hzz+eHnzLF7wMNZDry\nn/3M/vz556uHOhXdkdZcM7PRpIapUS4LbrVEwBtvpE3ThzPPzM4NaSZSz0AB4OmngRdeSJ+uDi5h\npO6LzZIwXO/76KPm9FrRrbZpsFVwaCWNGZPtiaRD/6gvvgicdZY7LflRGzHzeu+99Gm2oq4/hTcM\nJ42ZbFeNNOjuBq67rvh8XPWSWl0DAJtvnvWpotHOEkYKGwZQLbvJ4aOUMJiIYVWThKFX9rvv+tPh\nNuJWQCuchyEEcO217jTy7gLKTaMIlWUr2wCAYggD4B8KlAdFLNwjql9At9ZatVvc2PJthkrKRRil\nhMGErZLyzp45M/pG6na5eTz6aPgqUd+9VPjoI/uq4ZQqqVaF/g0/+KCxJFOESgpoDGHINn333W4V\nNJcwJFHofeXNN9273UrEEgY3vuu5iTA+/bSUMGpgq8CiTtDiSBi+U7NSgvueX/oS8LBhZUuqBXdF\nLdzjppVaenj9deDAA/nh80Cvuw8/bEy+Ej1BwjjrLOBHPzKHue66bMB3Qb7jJ5/UphtbHg5MKqk8\n6zBMhHHJJaWEwUKKSiICnnqq9t7Chf54po++9dbF6KqLmIm20mw+VLfLVUn50ps4EbjxRn/5ikBR\nM34b8npJ2dBICcNVZ4ccAtx/vzudVIQRAt8ka8EC4Kqr7M/1dEzjzsKFpYTBQkynS234UtN77DHg\ndtdSxkiElNk0MLSCfp0zyKcmDB+aWS8pCCOlW62si0WLsnacKt0UkO+Zt87kO8oJYaxEm9eGoca/\n6y7gu9rOei7CsN0rJQwGbB2+ESvDbV5SqVxD58+v/g4Z2LgzvkbbMDhITRi+ekvZyUL3MWo0WXEJ\n48YbM0mZi1aRMDjgSBimAV5HSpWUSWJQ05eq70bvk9VWhHH++cA111R/p3KrTYkivaSuugoYMKD6\nuwgJI5UNQ+Lxx+vvLVkCfOc7/nR1Inj8cfOutKlUUq2wGK2ItpvHrVZi2LDsP/eQn0baMBpNGL7y\nhORpi++buHHql2gZJ4wf/Qg4TTnPLyVhpGoQRUoY06fX/s5LGCakkiZc6cybB/zxj/5w+sxrq62A\no4/2hwstTyuh0ZMdbn59+2b/586tvW9rg+0oYbSSSsonYZjSMaFUSTFQxAxx7ly7J4aKIiWMPOqL\n0A5smu00a9A1ieIqZJ2nUjlyOtm++6ZZ6Gf7pnnqWo/rSourktJJmWP0DjlKIAaNlDBSw9R31bGD\nqxouVVIBaKRK6n3mAbI2tYarER57LI9oYghD5hsqYXR0AA88wIvjSicPWtnofeutwL/+xU8ztD4a\ntQCU2y64RCHRuzfQv795y4pUaKQNQ0UeCUPm5VvpzZUwfFjmJQxOpbWCDSOEMC69tNpYXYiZLbiM\nYz496ltv1T4zreWILZcpP9uzPIRx5ZXAffeFrdVI2X5ipTPbep6Y7WBSbA3CXSsgISXat9/mhY9B\naglDLtyLla7zqqRk3Z59dnUjQRWmuvfthpxawmiFE/eSo4iV3ty4oTNDrreDKQwnjqkTuKCGW265\n2nvPPcdLIySPFOFdnfroo7NdhDfaiJ9eyk7mG3BsEwtTO5o/H1httbQqk1DC4ObtWh+QCj4JY/x4\nXjqyrLIvFkkYRFk412TtnHOA3XazP/d5/JVutRY0UiVVFGHIPaxCxFnbbxWXXALcdJObMHwSRp8+\n/jKlgktn66sbWee2uhciXl3ACc9BCgnDt6NyHh08d5Vx6DcpcucDn4Sx//68dHRyK5owAL9KyvSt\nQ9WCan6p0HaE4aqADz7I/rcCYXBVUtJNMfVM7PjjgVNPraYfI2HohFGEMVCmedFF9md5bRg6Yahr\nWYQAPv64Nj63k4XYkFJIGD6cckr9vc03N6c1blx2sqLMz+YuGythyHBLl2b2vyKOok1twzANyC4S\nyQNTmdXvZNt9Vs+/9JLywPax7r23ephQEW61eSUMW3w5k+AMEPp7+cq8wgpVCcO0fkGPv9tutfvu\nSJVUDEIHlhNPtD/LSxjd3bXPttsu2y8KAP76V2DgQF5ZdRRJGDEzc93lFcjOpjDZx8aOBX7xi2q7\n03do1csXShiqhHHvvcDPf86LFwKbhPHgg2HpNFIlpedpy0OOC/PmVe1WMYSxzEsYNqgDYsxK71QI\nlTBCCCPUhrHCCtWZCufQ+gkTag3beSQMrsdSio4YKmEAVZ97k6eTacLx2mtx7SfW6N3dDVx/fa3R\nWJ/5qltux8InYejhYgijqIWQJgnjnXeAbbcNSydUJSUEsMUW9XvLhaikfOpg+T3GjAE23DC7loSm\njhUuG8Yyv3BPh4lp77+/uSopm1qjCJVUiIQRA0kYMYNejApMxTvvVAdErr48hDDku6mn7LlUUuut\nB0ya5C6HCzESxsEH127braZxxhnA8svz8nZNoLh68VjCsLVrqQbMAxNhxLR3E2EceSRw8832937q\nKWDOnNp7eW0YJpXUW29VJ8MxNoyWVEkR0a5E9DIRvUpEpxmejyaiR4hoERGdHBJXh6/jbb99c91q\nQ1UJIUbvUJXU8su7O1CRRu+8hDFkSHWhZBESxj33ZB3QtAOxrf3oqp0UKikbfBMIjscaxybjyyfW\nrdZl9J48macGXLrU7elkIow8Ey91QL7mmswl2/TdXO0sNE81Xiqjd4jKKhS5h1Yi6gBwCYBdAGwE\n4GAi2kAL9h6A4wGcHxGXWY7qdSsYvbnxG6WSajTyEgZQVceYOsCHH9aTbYiX1PHHZ/p9U1ux1avu\ngloEYeQxesfkHWsn8pGRSyVlsqeZ8NRTbk8nkw0jT73ltWFw4FJJmSQMNZypfO1o9B4DYKoQYroQ\nogvA9QD2VgMIId4VQjwBQJ/veuNyoVZcETYMm9599mzgmWfqn3GZvZVVUtz6Cp3FpLBhrLxytkJe\nfa6HkwZ8E2EAdh0vlzA4CDV6S8jBIWZ7DdvgLqHumeSbtaYwetvK5wPXEaUICUPPg9NmY1VFpjxc\nXlJcG4bteR6kIIzhANQ1wTMq9wqJy+l4RUsY6vUhhwCbbVb9HWL0PvTQzLCpxnMhRiWlNrxYaaMZ\nNgwVtrp59tna53pal15avW/Kp6MjTMKIaVd5JYwrr6x/FpqnHm/cOH6aRRAGF9zBLpYwbrwR2GWX\n+HUY+r2URm+uSsq30jv1WNhmK73HYuHCzCUQ6Kz8ZWgWYeiDsPyov/41cMEF5jgS//u/2UxZjedC\n6Gyho6NWwliwoJqfLb1U4jhXDRIrYQDV2betnOrgayOMEAkjlLBNZTHh1VeBUaNq7+WZmUvYiFSu\nV3KFseXNbRMulZQt7k9/CsyYAVx+efabU7866YfU0U03ZbYsec6HSeUTYmiOJQyTGtKkkuKWZcGC\nSQAmYdq09FuzpCCMmQBGKr9HVO4VEHcs+vXLCEOdJQHFimE6XAwfKhJzPYFM4Ijsanl0wvAhlQrP\n9qxowlDjhxCGbcJRhErq9deB0aPrw7k2n3Op2Dh5q6v/bWG22642zUZIGBddBLz7bpUwOBM/fRY9\ndSo/PzUNoH5AdtWPCUWrpLg2jP79OwF0Yv31gcGDgeeeG1cfKBIp5uJTAKxHRGsR0XIADgLgOpBU\nfcXQuLlUUnrcDz/MXBcBf8Pm6qBDjd6SMIpQSekDizqzVONfe60/b5keFxzC4KTR0WEPL1ds+yQM\n17czzfRSGL0ffzxz8faVwbaWwiVh6Gntvz/wj3/Uh9cHQLlYUR2MbLNW6UKcd6V3HiM0V8JQw5n2\nYOLm4yNp/V4e+KR7vV0sXcq3YfjyyYPchCGEWArgOAD3AHgBwPVCiJeI6BgiOhoAiGgIEb0F4CQA\nZxLRm0TU3xaXk6/LY8g3M5ErQZ95pmpDOOWUqk7chxgJw9bIXP7qRLXbqod6Sen52s5vOPJIc3hu\nx/A1ftszjoRBVBtm0aKqMVsnDL0OU6mkbHYpVzvbbbfMxTvlQGtLa/z4ajt2hT/qqOy/qqYsigAk\n4YWo1vT7nPadR0+vTxB08lTbXrNVUjff3BrrMJLYMIQQEwCM1u5drlzPAbAmN24M1IrxVdLLLwPb\nbFN/37WYyDaQ5lVJSTzyCPCZz9TfX7CguuWJjlAJY8YMc3yuV1lqCYNDGLqEMXEisNZaWZkXLcru\nxSzck2mb2op+z7Yfl6vsLr09px59EgZncNXrRRpSORKGK28OYrykYgjD9g1DIPPRVT42lRS3/K68\nfJMsfRz59FOzSoqbXyq09UpvFTFGb73iuYO9Gk/mNXy4OU0uvvWt2k3xTIgxuqrl0VemcpFCBDel\nF0MYEuo+V7EqKV3CsKmk5Iw8ZDLgUuW4Bh/XTN6Vv6nN63lLwlAljKLcamUeIVuDyLLMmZOdXV+0\nhKGmoeZv+j6m984zofKt9NbLRhSukrK5jedB2xGGT88M8BtQyNJ+n4QhvRFCVVIqTD73LmkmVMLw\nbTDnyjsUMXVpgq6ScqUXY8MIkTD0b+vrrLYycIgnxIYB8AhD6sVVCSNUJcUNL8kpRsIYOjSbPOnu\nsqb2GyJh+Mqedx3GvvtW+/Ds2eYwuoTxu98BV1wRVr6vf70+PRtaceFeUxBiw5g7t/6D33MPcOCB\nteFcHZk7yOUhDNM2FXnAJYzQ2a+OomwYQriN3qb0OPcluAv3iiQMm+pKDaOnwSUMm0oqpYRhqwNJ\nTqlUUkccAay7rrl8eQnDppICwuwG8+ZV9yYbNixTfdsg8zzxROCOO/x5CFEtn7qnWdsZvVsFLrfa\nwYMzLxL5QaZPzxbs6MgjYUgUKWHk9ZLSCYMjZsciBWEAdsIwdWqX0dsGU7uJJQwXEcycaS6n7f3z\nEIYtvE4Yav0LkeWpn13/7rvZvkq2NG1lUre84Q5aLsJ47LHarfcl8kgYurbCJ2GEHo+r9+cnn6ze\n852HIWFahxGCZV7CsKmkfK60775bvWdzoY2RMPR81bRDB1+fhJHXSyp0mxDOQMXJ15WuK9zcudlg\nXZSEIYTbhiF/c7ZvefbZWrdbPY3NNgP+/Gd/OjK8qR2pA5r+ThyduJz12ySMyy4DVlmlNs4zz2Re\ndKF1rNowdHAlDIkjjghXK4aAQxhAdn4IALz4Ylw+999fvXapEE1lU8tne25CKWFYYKvYkFl0jKrJ\nZiANSVvCJ2H48vbly5EwQp7HxuWq9958MxvkYgnDl4dUefnSsw1+av3rK2p1wgCqNjPOTFENw1GL\nmFyEbRKKTW6sAAAgAElEQVSGzYZhmsHb0nLd79u3eh2ycM+WtmudkI0wQvo9V8KI3b1Z1rc6IezX\nj18+CVNd+rzjSsJgwDfL56gBXHFcH9bm416ESsoHn0pq1135ZQsJZwv78MPZYskQlZQtLVun5saX\n910dSrYHjoRhUxea6j9UwjCtQPblr8aT/+UusapKSk3bNbibpBxbmVRJK2RrEJtKUS2zjpSE4dut\nNvRAMfl8ueWyFejSDRwwE0YqlZT63UvCsEBtNDbCiB1YALvUojfW6dN56Zlw3nn1J8CpadhUUuPH\n20X/UJWUq75s9aeX68ADs62pdXz1q8A556QhDBW2GThHlWZSSemDo48wJk6s38rB1FFN6XAGT1P9\nu1RStnhAdqaHScJQycMEPa33388MvKY81Os8Kin1t75LgYpGSRj6kcW+AVxN5/33a89TMR1+xW3n\nKlwOJ6WEocClzzcRBme2H6NGcX2QUAnj4YeBn/zEH07Pe//9zavUuYShhvnnP8331d8/+hGwzz72\nct14I3DrreZn3d3hK1a5HTOUMLgShs2oLuPuvHO2Etf0TM2bI6m4JAxXvZmcA0zhdfdUDnmpz9Qw\n22xj719yQEyhkgKqxvipU2sJL6WXlIswZs2q30HWlJ6tPfXqZT5bXYWvnXPrUj3KNTVhtNlutXb4\nCIMD1wezdSzuB7n3Xh7ju8R3/dmECVVjPmf/LBdhhNTTDTdkR0e64CpPKJFyRf9QLynb97CpX1wq\nk5D65xBGCrda22BuU0mFSBhAlo4tj379MhVMnlmu6T1Hjcp2gj5ZObeTK2HYwnDcanfYIS4Pid69\na1VSXAlInXiEqqRKCQO1OnAVasX4/MxTShgu6OE4KiH9A++3n/0ZANx9d/Y/D2Ho4Wz35G+X8U/m\n4Zr1pVZJ5ZEwXIONvhuuq8PqdSLTDZUwJEJtGKY1JTLcnXfWpmUjo1DCUKUaE2HYyhqjklKhb+HD\ndVHlShgffWSPs9Za/vRM0CWMkHqR4L5TacPQ8O9/h6uk1Gvbh+FKGCeeWL3mDIwStp1JVejvpZ7m\nZ/r43/qW/RlXJQXwGqP83VuTS9W85cDJUdXlIQzTLDCUMHzlHDy4Nr6rfbjqRMIkYchwd9yRLfyy\nDcD6PR3crSb0Os8jYej2n3nzqmlKfX/IwOiS4Fz3TX0wxHZy3321+Z9+uj3Oxhv70zOBQxgxKilX\nOqWEUYFJF+hzqzUZN3W4ZiDqM9XNj2vDAMynaIUgZjuUPIRhS4vjXsgl0iIlDPW5beDKM6CpsBGG\nScIwedLtuSdw4YXmMpgkDM6g6XtnG3mYYCMM+fv224FVV62GUSUsV7v9y19qT0b0lV+HzYZhauu2\n9KQEJgdk1f1Vj6N6N3HKJ+u0o6N2whiqklLT4uZZShgKYozevpmi7X5HR9wqSx0hEoav8eiwzTC5\nhPHQQ/X3YlRSEo2UMPKopHz5339/rRHRBkkYrkFPEobrqFyXhGGSpF54IfvPVcvobcK1GNBULjVN\nVSWl7p2kE4YLxx5bfza7nqcPeSUMCdW2Y4uzwgrVa44zhrrTsUz/0EMbo5IqJYwKfIOpiTBMAyIn\nXZm2i0xsWLoUWHvt6m91u4SYsnD2LlLjzpxZu52BizB23NFcBrk3jopUEkYewrClx7lvS9sUfvvt\nM7dZgGf0VmeUehgTYdi+qU/CkJAqkhgJQ0+Tsw7DJmHoaboIwzYgp5QwYgjDtI29HkddlMjpo6rd\nSrXvcVVStnHNVYZSwtBgYk4fYXz4Yf09Ha7V3Hqcn/60Pl8dixfXrsuQhNGrl71zxkoYKmTcyZOB\nQw6p3g/dGgSoPfJSpsvR1zeDMFzeULb7IWsiOEQv69j0HeXWIK7FnS4Jw0WAXBuGTcJQr11rSGII\nw+X5l5cwAD5h6NDT56yTMW2rr+LHP7anqbaNIiUMub1LKWEwYeqQnIEhRMKQ6yU4fvwSPh2mml4o\nYdgGARUxe0mZOrROGCbksWFMnmwPb4LP6O0iEr0sW21lPzeEMwhx6tilkjKB4+nn2hrEdY+rkjKd\nPW8iDPmfq5LiSn+2+6p3mK8PcCUMFXo6Pglj/Hhzmqp6irOhpukZVy0+YkSmOislDCZMEgans7vU\nTtzZjq0cAM/oHUsYHEJMRRgxKqlLLqlPxzRILVkCfOEL9eXwldMULoYwHn8ceP55ezrvvAMcc4y9\nLEuWAGecUZXM9DJdfDHwve/Vp+16l9iFezZ1RYxKSq4h4BKGqx37JgycZyao37lZKilbmo1WSfXu\nXYyE0bYL9/SKNPmtA9nRpwCvkYZIGBKumbQ+QHM20uPaKXQUIWHo6foIY+7c6rVe1uOPr09n8eJs\nF1cVMTNDH2GErsZ3OUXcd1/1wBsTliypfa6X6Xe/A157jV8W9ZqrklLLa8rD1FZ8M1g52bERhpq+\nPuvn7iVlKqsPahnU9SZFEYZPJWVLs7u7WqaiVVKyzltWwiCiXYnoZSJ6lYhOs4S5iIimEtHTRLS5\ncv8NInqGiJ4iosmmuDrkYjWJL3+5upEeUFuJu+9efy9U1237wPKZDS4JI0ZU50oYKVVSJtgIY489\nqtcclZS6QEoipFPo15wZtv48pOPaZucqfDsCu6QztTy2Qd1WblUl5SIW9R1sEqQJJsKwSTU2NZEJ\neSUMtU/stVf1Og9huKR1VcLo7nYfSaCqodRrWz/2qZxMz2fPBtZYoz5fOWa1HGEQUQeASwDsAmAj\nAAcT0QZamN0ArCuEWB/AMQAuUx53A+gUQmwuhBjDyfOCC2p/P/aY/9hJjsrm17/O9t43wWUQt8FE\nGK7OrKbnIozY9wsljOeey1Q0eh42o7dcuKXe06EOUqY65XZ0eW/SpDCjtK0sapoxhCHzPvfc2nrQ\nwd3xNFTCIMq2bNHLa0rXVO5Ro+pXUKswuRbbVFI6ebkkDO5iWVt8029b3CJUUvpYZEuTo5LyTRJN\n7/Tqq9k+V3o6Mo+WIwwAYwBMFUJMF0J0AbgewN5amL0B/BkAhBCPAViJiIZUnlFMOVwbecUOqI8/\nbt573zYTBdwfxKSS8pWBMzPzvZ+to4UaW087DTjrrPp85aCnDxbqYM+RMEzlNG2i6BpUttsun4Rh\nahcxhCHf/X//110GnTBsrrqmQdVlwzDBJ2HoYfSBR4Vsyy+9VL1nkrylGoQrYUhMmGAuqw826T+P\nl5S8P3Ei8Mtf1obRVVIuu6QuYeT1kuJuPtjSEgaA4QDUrehmVO65wsxUwggAE4loChF9N0F5ohuQ\nK70UEoZqw4ghDJ+qAQB+9jO7mDx/vr2sIdA9YEwEwLEbmL7Jl75kD+9Lz0YYeW0Y6n3bYCsHA52U\nfYRhSy9E8gHMdWmbZdvsFq4+Ip/ZNv2TZTv//KwOQo3epvU+MX22o6PqIeTK1wSThKETmS5h+AZk\n2S5efbVewvjmN2vDxqikbPUb66jjQysYvb8shJhFRKsjI46XhBAPm4OOVa47K3/1iJUwBg4069VD\nCGPffYG//S271hv8AQeYG6UpPVdHczWcs87KXENNmDcv8wBS98MJgW8Gn0LC4ORrex4jYbgG5mHD\nslm3tJl1d9vbEZcwdHWercwmCcNFHCbpkTOL5aqHTOpMk0pKSqRclRRnEuSD/g62dU5CZCrMzk5z\nOqa+qZdBlzB8hCHTlHu+AdXB/P/+r758OtR65BLG++9PwsKFk/D227wFyyFIQRgzAYxUfo+o3NPD\nrGkKI4SYVfk/l4j+hkzFxSAMO2IJw3Wfq5JSTxtz7WeTRyVlW5DlQkdHtup7k03iZx562V0SBied\nVIQRqqpR07VJGFtuma2UB6qedq5Fb3KA8tmJUtkwbOoUX9qcdzDBdjyorz1/+KHbNhIqNelxZs8G\nxo2rLZNtK58PPqhVYeowGfb1MoR6SfnqTUWMSsoUZ+DATgwc2InVVwe23RZ46KFx9YEikUIlNQXA\nekS0FhEtB+AgALdrYW4HcBgAENHWAD4QQswhon5E1L9yf0UAOwPQvODd+Mc/6u/FEoargXIlDHUG\nGeOp4QrHnY2ZnptO+AqFbzZsc21WsWAB8MUvusOY8t1mG3+5UkoYvXrVl8+mqgKqA7b+zUNVUvL3\nAQfU5+siRhNR+YzeISopU/outYfsF3femZ3AaIPrG02aZH9mg5ywmd5F3Sbnww+BF1+sfW6SMPST\n/nQvKZ+EYfoGIeswQp5LdHVl7awIt9rcEoYQYikRHQfgHmQEdLUQ4iUiOiZ7LK4QQtxJRLsT0TQA\nCwB8uxJ9CIC/EZGolOWvQoh7QvJ3NUYVHMLgLvBSEUoYvtkwEXDRRcB665nLYSsn5/3yIkTCsNWl\nal8JMeK5RGtbvZjIwBZXvRbCPEt1zc5n6jK1JV+X0dtXRq5KyjexsNWLa0AKmSnLZzZwvgtQ3X4n\nBNJDy1Re6b0mBHDmmbXb3gC8A65UwnjySf972tRMXAlDpn/GGcD66/PiLFmStTOOyiwUSWwYQogJ\nAEZr9y7Xfh9niPc6gM30+/nLU3+vURKGqpKKIQwA+OEPgc03tz9vVcJQ39dWlyqh5lFJcfXgAM/F\nV4XUg+vlW7rUXs9yg0JXOYHa9iHzcoVXw4SqpELcam3hJXzH++rl4RJGavTqleW95Zb1z6SEMWOG\neXsbjhehqpI64QRg7Fh3+FQSxpw5wDrr+MsHVAmjJSWMZoPL1I2SMFRjr0tNkGdmxh1Ai4aeD0cl\npdZP6LoJG2x1qv421as+EQiVMGLq2UUYvnbpyi+FhFGESsoHH9mHwrSnloQkjJEjzYRhsmHo4Gy8\nqSKEMFwShi0tE6RKSvVWS4UeuZdUSsL4/e+z++qurypCjd4pCCPG6J0CITYMjsEyj6uzqxymvGwd\n1BTGJmHEEIYeRvcec5VZv8dVSUmE2jBcBvtQo3eREsbs2cArr5if6YSsQq0j07v6PBiB+u8XQxhF\nbw1SpITR9oQRI2GEpOWrcJeEEWvD8IGjklJF51RIYcNoFmH44urpFEEY0oNHT9OHWKM3px+oYVyE\nwXWrVZ/ZEDNYqrjqKmCDDczPXBKGz4ONc+Z6CsKINXpznWi6urJNMqdOLQmjDilVUib4KlxvQFzC\nCMG4cVlj50oYQtQeJZkK+uAwY0btb45KSgWXMHzfwFUvvng2CSPU6O0rG5CpM/QZcJEShk+q0jfV\ndOnwU0gYjVCVqqvMdfjaW6MkjKJVUkuWVM//KQmDAZ/olpIwQiUMUwe77LLqClqZnpru2LGZF46M\n45tNNoowdtqp9jlHwuC6cargEoZt8OUavX02DD18CsLQid412VHLp4fTB/tDD7Wvmo9RSdlsGDa4\nCCOvhOGCq0y+78UxeocQhu17NkIlJTUMJWFoaLaEoSPGhiE9O1z5qfdN+9fo7+c7syKmg+px9EN1\nirJhcGdxoSopH2GY0oudeAAZYejvEuNWa4I+2Ol7WgFVicL2DnkkDNMzHY2QMFwqqUZJGDZCVtOI\nUUlx608avW3ly4O2JwwgXO+fsuG6BgCuDUNt5C7CkHF8hCHTdJUzD2HI/+qeUn/4Q34bhm2XV9es\n0VQuvQyudzW1C5vRWx8AQm0YoYsB9TCu9+HOjl1SkiuNFDaMRkgYeVRSS5eav5GKIgnDp5LixpFG\nb1v58qAkjJxpywYkw6lnhzeLMOQM2YUYW4peb1KaEgK48MLaZxwJQx+E9POQJWJVUj7CsKkMhLAT\nRqiEoX9nF2FwJYxYwpDvZFNtFW305khKeZFHwgCyOgqRMEzwEQYR8LBh86MYCaNUSQVCiOYShnwu\nB09180KuSopDGGpD5RAGd5DNA1XCsA3Wrnz1MDbPrkYRhiph2GwYpvQ5ZQPMM8tUC/dM59jrkIOh\n7Rvkcas1PdMhBPD//h9w0knmZyGwDdz6ffVgtSIIw0WM3d12CWPBAns8G0JUUiVh5ERRKikJuaBH\n3WjNZZw+8sjqPdNpaTp8EoYer6MD+NWv7A0nj4Qh/6v2Gp+qxVROvQOvuKI5DteGoedpOvTHVB5V\nUlLvhUoYe+5Zn/73v1+9Ng1GIaTDlTBchMEhKBP++c/6ezESxgUXmAfLUKhbdOj55pEwevd2fxM9\n7ViVlAk+lZS+lYktTqmSciBGwuCkKcFVSUmvJK5K6p13qvdMxlAdo0eHqaSIgJVXtntq2QaKlVay\nl8Flw+AOhK4ZbSxh+CQM3dCrxjMRhm3Bk8+GseaaqIO6dxZHJVWkDUOXMGwDGhcqYejpcOpPfxYC\nG2GkUEm5wCEMqZ61fU9b+WK8uFyTA1desWh7wgDSq6RiVFaDBmX/f/ObKnm4CEMd9Pv390sYquRi\nIwy5+lXaMDo67GUwNc6RI+0raE1liyEMFXrZbK7AsYQh0/eppKR0KMMvXWpeZNfVVTvT1t/RN9j4\nCMMGE1G4VFI2fPghcM454XYYF0IljFTSvW335aJtGDpMef3hD9n/FBKGryyTJ9ff69273raaCm1P\nGEXYMNSw3AqXhLHuusBuu2XXXMIYMIBXNpeEce651dWvS5f6O44pn1VWAVZbzZ+/rpIydYwYt9q8\nEoZtMPZJDJIwZHhZLr1j33wzcPvttfFV+AyiPi8p24AaopIaNszdhv7xj9rJRx4JQ5dWVIS61aaS\nMFxtJQVhcMaDlCqpOXPceb36av091aW+JAwNeiWbFisBxRGGfL7yyvXhXUbvxYszyQLITvoLaYgm\nwnjiieq1bYasPrcNFCHlMB3RKRGzcC/W6G1T1XDPddbPUJD1p+ernyOv58eRMGIkMZPRW4ckjPXX\nDxt88xCGq0wcCSOPtOEyere6DcOlkkohgakq7pIwDJCVsuee9plHyLYVIR9NNlwpYcQQxoABfpWU\nCtfB8zIPqZIyQd1mJAS2AdI0M+Z0TtOZ5ybESBhLlvjVNLpKSpUwbCopFTEqKRdh+CQMlw1DqhJ1\nlZMLKVVEMYTx9NP2+LFotEpKtpGNNqreU79TqISRijA47r8xaHvC0FVSsQYlW1jfYCVnxWusUVsm\nwL01yOLFVVUUlzDkM/U0NhOkCsZWL7ESRghhxHhJHXNMeJnUNNW0R4yotTdwjN6qhAH4CSO1SsoG\njoQh0bt3a0sYKrbYov7ZGWfw8nW1iTxjQKhKKqWXVN5vIdGnTylhWKF+XK5LXUqVVJ8+wE9+Auyw\nQ314nw1D6uylpKHnrYM7EPgkDJsNg5OuWg6X0buRW4OYJAyf7lfGM0kYknD1+tMlO/3QpBQqKY4N\nwxYOcK+N0PHss/aDn7hIZcOQdjdfHYYitL35CNe2Y4Lp2/gkRtP9VBKGREkYBthm0mrFcRY2SXCZ\nXjbAvn3NXglqA91449q05eDzpz+ZCSPPh5adz6WSSmHDUDt3CgnDhth1GEB2Fvjw4fY0bCopU74+\nVWCMlxRnIhMiYYQQRl6oA5z+LUMJY8mSrP64qhQ9/V/9yp23qYwqPvOZ7H+oSsrV9mwShqtvpFZJ\nlYShQVVJudQwOmFsuKE7TVMaOtTZiOkD+WwYQgCbblqbp2tQ4jYmm9FWfZ7ChqFKGNwFYUVLGD/6\nEXDLLdVnRx9dP7MfMQLo7KyW2+ZWq8NXXh9hxO4lNW9e7Tez1e2GGzaWMIB6kpUIJQyfVKwjRiXl\n+n7SsyhUJWUiDJ9KylaOVEbvlldJEdGuRPQyEb1KRKdZwlxERFOJ6Gki2iwkrj//6n8uYay6atWz\nSUeISkoSFneA0QlDj3fXXe68OOCopFKs9B4xovqbq5JS81W3UXGBu/lgdzdw/vnZWhgJ0wAgv5du\nw9CN3qGdLUbC4KiZvvGNbAatShqmsJde2njCWLQo+5+XMOTGf7EqKdUGGEMY0hYZSxgmpxobYfjO\nR8+LlvaSIqIOAJcA2AXARgAOJqINtDC7AVhXCLE+gGMA/IEbN7w85mudMFwDQuiqcFXfrabZ3V21\nU8j7Lk+XVLDp4CVsEoavTLqtYOjQjHTHjq1dua6GceVx9tnu/CRCbRiqH7pNJ60Shi5hvPwyL18d\nLmK77Ta/DeOdd+wLJ6dN80sYffq4z9kuAtKzjbvS2wZJGLHePZxv5fKak23G51arw0QYPgnDtGJb\nbY8urLeev0ytLmGMATBVCDFdCNEF4HoAe2th9gbwZwAQQjwGYCUiGsKM6wVHwtD32uESBqfC1bTU\nGZ4Q7iM51dkQp5OHShi2srtsGNz0AeAvf7HPCDlGby5C12HohGEazNR61yWMefOAJ5/ML2Go8ffa\ny6+SuvFG4LDDzGl/+qmfMOTMMpW3jQ9CNE/CSO0lJScNoVuDmFRSPgnD1Dek9OnrH3ffXStBm6Da\nblMjBWEMB/CW8ntG5R4nDCeuFzaSUK85or/pOXd2a5oZyYFbTUcvhy1903qSEBuGSyUVqyvVZ/If\nfWTvYBwJg4s8EoZNJSXjqYShdmTbeh4XXIQBmLdq4Q7uixb5607OLJuhksorYYQavXXkVUnF2jDk\nlvwrrFC95/OSsp1eyJEw+va1b4sioUrVqdtCgVzkRKSgNFa57qz88SSMugI4nsWopEwioIkwuBIG\n1yBsgm+lt23mwx2Y1bK61F6uNEIQShjqDEt2Hn0SYFJJqXXi65Qm+La+zkMYn37qV2f27p1J0u1s\nw0ix4MxWpxzCMKkw99kHuPVWe9xddslIQN3LTZaDq5Iy2bdM4NjWFi6chA8/nAQAuPNOf5ohSEEY\nMwGMVH6PqNzTw6xpCLMcI66Csca7oYQhGT6lSkolBnVg9Z3hbLpWy6gilUrKRhihNgzALmHMmuVO\nIwQpbBgmKUP3kuLMkl3wSRi9elW/6/HHZ6ov0+ZxJqgShm0m2qdP/Cr+GAiRzoYhJQxunevh1N82\nYuBKGK53MZVvwIDa0yJ9Now8EoarX0usumonVlqpEzNmALvvDkyYMM4dIQApVFJTAKxHRGsR0XIA\nDgJwuxbmdgCHAQARbQ3gAyHEHGZcLzgqKRUTJgCTJqUhDPmBbRKGfl/XdfpIzZSXDyYJQ1fPmdLy\nSTUhhOFLIwS+b/DGG9l/m4Ths2E0gzBGjMjW5oRIGK1mwwDSSxh5jbS2QRoA/v1vezyXSso3puj9\nyUcYNhsGhzA4Eoba/lNPHnIThhBiKYDjANwD4AUA1wshXiKiY4jo6EqYOwG8TkTTAFwO4AeuuLFl\nCVFJuZ6HVLLuJaU/c6mkurrMZSDibVdtg8+tNqQh6/Fk+hKpV+Zuumm995TvO8rwJqO36ZQ5lw1j\n6NDqs5ReUkDWkdXvSsRXPeoShikvacPIo840wbYppCxX375pCIMze7alr35T9QwSFR98YE9Pthm9\nXvfe218m/d1jvKS4Rm9O/fTpU5yUmcSGIYSYAGC0du9y7fdx3LihCJUwfMizDkNVSak2DDW8xOLF\nZhtGnz7ufag45XfNREKMcbb4EqE6Z987dHf7Z+quuIDZS0rd0lumabJh5Gk/vnJLG4P6PMTo7Vtp\nLiWM1ITRu7c970WLMoNvKsLgtqfXX7c/cxGDDTYbxgUX1O5vZXov29qaGJWUDy7bpESre0k1HTFG\nb8D+gfK61arpuyQMlTBUyJmobnhN5SUVK2GYXAhDJQzfAKkO4BJ5CKNXL+D992vVEbrRW5Uw1NXr\noeCs9NYljBCVlJw5u2wYjZQwpk0DnnsuHWFw1C0SOoHZnEe4sKmkOGOKnqfP1rTHHvX3lhmVVCsh\nlDBsCFVJAfaFez7CMEESRmz5f/KTTK8fuo1yjEoqtEFyDOupJQwdLhuGej/UU8rnJRUjYXz729l/\nnTBM8YqSMGR99u0L3Hdf9f6772b/hw6tlmerrYCLLjKnk5IwUsNGGByJR9ck+NZhDBsGrL567T2u\nSopTHrX9p0aPIIxYkuAYvTkDnK2hm3TNNglDzad379rZLrcsKubODfeS4g40atxQWwtHJZVXwlAJ\nx0YYsiwqYeh1rm4KyUGISkq2Gx9hqGnKA5waTRiyfj7zmdpt/AHg17/OttmReXZ3A+usY/5m669v\nz+Oaa/IRRl57pHxHXTWk21VM+UgPLx3f+IbdG9HmtZdawkiNHkUYemXG+nSHEIa+DYea/wUXxKmk\npHuk/iyEMHwqqTwShhrXZMDjpGGDiTC4kHWrznBtaiLZQVWbiS5h2I6MtSHES0rCVx9qmhyVlO0c\n9zyQKimTV5/umeXy/NNPLDQhxcI9ib0D94yYODGTovIShhq/uzsjDikpyuf690u5DqMkDA+KNHpz\nBjg1Lz1Pl5Rg85KySRgh8Bm983hJ5SEMX6fIQximb2VKS+7qKuPYJIxQwuCopEJtGDbCcEkYeTzs\nTJCEYWofcrGdutNvnkE/lUpq662zxXbc9IQAdtyx1mmFWyYfYYwalW2xr+enwuTNZwKnPKqXVGnD\ncKAIG0aohKHHcUkYtjzlwJJHJVWkhFG0SirWVVct16qrZv9NhHHDDdWyqDYTXcLIq5LSEWPDMBGG\nLJ8p/SJVUqNGmaUm9T3y2CJSqqRijeA6YaSQMEzvZZIwSpVUg5CCJFSYNhJzhbV5SQFuG4YavtGE\nwXX3M+WvryUJAccTK68NAwAGD7aH69u3VmesGr1VL6lQwvBJGDFeUuoxuC4J4/nnq+3QVMcHHugv\nvw1Swhg/3vxOqoTxwgv5Bv28/Ve20dh0evWqrT8OYXR11RKGrqEw9WObSiq10buUMAyw2TBSqKR8\ncHlJme5zVhNLG0ZKlZR6nVLCaCWVlJq2HORs5VMJwyRhAOEqKdssV8IkYYQcyjR/fvZfJ4wLLwQ2\n2sidZp6BQw5AJq8xebqbSoQhC/BU6JJ6aFwVsSfO+dxqVay0UvbfJ2FwJn66ZGNDqIRREoYBqY3e\nm2xSveaqpNQyhKikbBJGXi+pIlRScn+oadOq9xohYcSk3dEBfPObmQujCT4bhnqfC86CKp0w5NYa\nNqgSxuuvA/361X+/ww+vTbMowpDp6+XTJaVmShh6GbjvLcPp39wlYch6PuAAe1ux7etWpEpqpZXS\nE5HAajsAACAASURBVIVEjyAMiddf50sYXD25b0bs2xnWN+jbjN55VVKuMtlEX66EoXq76GqszTd3\np9HdDXzuc+7nsTYMvSx//at7LYUQ2TvbvIBCy6HG3X57YOeda5/rs9cXX+SfE750KfCf/2QH6OgS\nht7mUxOGunDPp5KSYZplw8irkgoljNVWy1aCh0oYJqM3d2sQ07utqWztajtJNAV6BGHICnz5ZX5D\nSbHRGcC3Ycj7HINkCi8p3261MYSRAjErvbmw7dFjgpQElyyp9QJSJYzQ+lfr+7bbgNGVDW9eeCH7\nr7+XXPjmghyIFi7M4ks1ma42Ua/V73jssdXwY8b48zMhVMJoBS+plIRx003msKoqykcYukrYpGmQ\nExgXbHXUr1/1WiWMUiWlYaWVqhWoHk0IuBsNt1H7Klyf0cTaMNR8ZAfl+uybZsKtTBiu7xIrYXR0\n1Kt71P86TIShly0PYRBV32PgwOy/PhhdfLE/TRmnq6t6LQcck55eJ4xLLqnGeewx4IQTeO9iKoPt\nWUoJI7UNw4Xrrqu/Z1pLo6oN1XxUwlDrSHea4dSHVEltt507nC0t9QCnQYNKlZQRt92WLbGXFbjc\ncmFGb06lcr2kfG61EtdfX/vbppIC6s/JtpXXdDqcq5HaCGPUKHP4lHj2WeDpp+3PYyWMXr3sZyWb\noEp8Ngkjj0pKtSHJ+3p66qzQBhmnq6v2RL3u7mq59W3sbauLQ6CW9eCDge99r5q+Hk7mKSc6rWTD\ncOErX6m/Z5Iw1HPWdUnBJ2GEekm9+KK7zLaJoKp6lcb4ItDWhKEv45fukhKuRpOKgUNVUrfcUvvc\nVEbXXjCmmZNpc7hQCeO73wUef9yeb6OgSxg/+xkvXihhAG4JI4Yw9HxlfJmmaTDyQSUMuc5Cfj/T\nAO365ip8BzepZdtgA+Cyy8zpqzYME4GFopE2DJM3kekbjRoFnHqqOQ35rrZ3fvppPmHk8ZJSCcN0\nXGwq9AjCiJEw1HgucFVSNjc+Xx4mlZRthm3Tq5sII1TCGDgwOzms2dC9pA46yBzuiitqf+urqCU4\nNgwpoemOBnkJQ28T+gAVQhiSIFQJw0QYLs84Fb681eeusFLCUAkjVlLg9kkOOOmYJmY+UjcRpvpf\nx7331k/eTO7aebcGUUkiD2H70HaEsd9+1WvdFS3EhgFkZ/H6wPWS4koYHIQShqnhhxJGio5q26U0\nBDph2Mqli92mfZpc8Ysweqvo6KiXMPRBJYQwgFrCUCUMXSVlQh7CUMvAkTBi25LJo0jHDjuY78fY\nMEwL3Hz7gdkIw6VGVfN54w1gyy3zSRgmlIRhgSoays5jMnbq1zqEAP78Z39++kc8/fTa3yYbhgrf\nLNVlw+CGt92T9/feG/jXv6rPdtqp/sxtV12dfLL92S23ZOosANhmG3s4LriEode3bvT2QdX1q04G\neYzeevouCQPgSTD6zruq88ScObXp69cqfBME14DrIgzV6K0SWAxp+BwigNpN/FTkVUnZ7qkTCBPk\nc9e3VAljrbWy//pENKVKymZPTYG2IwzdoCxErS97iIRhgk4iMQv3VPgkDJNKymbDCJn1quL9lltW\nVwIDmdeHbq9w1dU669ifDR4MrL12NY3YBnrooZlajOslpdeDzYbhgrQLyLTyGr318qW0YQDVdqFL\niJw279thwDWz5qikVCN8HgnDF5fb/lMThi1dn0pKpsl5L5s24zjlrFIbYXz/+7VplV5SBuiEoetA\nYxruFlvU/o6xYZi8pEIII1QlpUJdpW4jMaDehdaVru+ZPjDGoKMjG3B0LylfvUnoq6h9IKoShiy/\nKmHk9ZIyTSLyqqRUo7dr4R4HJsJVoabDVUmpNpWiJAxue/Cl89vfmuvft49ZXpWUhEklpUv9EvoC\nUNO7qXaRllVJEdEgIrqHiF4horuJyOjQRUS7EtHLRPQqEZ2m3D+biGYQ0ZOVv12DCl9h5RjC8ImY\nvnASPgkjpuO4Gp+vMZg6t6kMnD2tOHmmIgxZhyaVFEcNFyphSMKQAx5Qv1gytdFbHzg49WWzYdgW\nysWqpFzqMhdhSPWTzc03BBwbhu+5rpKy9d8f/rD2XWw2DJtK6o47asNzVVJ6fmo+Npdazphikwhb\nTSV1OoB7hRCjAdwH4Aw9ABF1ALgEwC4ANgJwMBFtoAS5UAixReVvgi9DtQJWXLFquJTP8koY+ofn\nEoatITdKwpCLkLiEwTmDmfMsFWHI+LE2jBjCWLKkViWlSxh5Z2p62eUCPttzE9S60N1qTenEEgZX\nJWVKnyNhDBhQa5g1IY9KKsaJwxSGK2Ho7rQuD0PdGQcId0IIQctKGAD2BnBt5fpaAPsYwowBMFUI\nMV0I0QXg+ko8iehhZtgwd0NJSRhf+II5vOkAJbVMMj2bXcJUxlAbBlHtmcR62hwJw9XIGk0YsTaM\n0LM5JGGoKimbO2nMAKTXf8weP2oZ1IGnuxs499z6fE3lPOqoqmOCKV1TvBBJi8i/DoOj4pMeh768\nOLC1fZe0BPAJQ1cz/vCH9rKYJoAhBOf6NqYwrWzDGCyEmAMAQojZAEynEAwH8Jbye0blnsRxRPQ0\nEV1lU2nZICtG2h1C/LhDVVJbbWUO75sVyfQuvdTcYTiNVi2LT1JR/fZdeaRSSakqnSIIg6uzNp01\n4ctLN3rntWHo0OvFRximMyvUule3BhEC6OysTV+/lrjySmDffatxTeFMXmcSvu9hcqu1hdOx/PLA\nSy9l13klDKKqHcCmklJPvgvte6ZyyLox7bYgwVFJuQZ4n0pq9dVrf9tWnaeAlzCIaCIRPav8PVf5\nv5cheGjxLgWwjhBiMwCzAVzoDj4WV189FsBYnHfepP/O5uVOjfqAKj+quv2zDzYJw9aQfXvFyPur\nrQZ8/vP25xyVlGtDQpOHRSqVlJ7n0UfXPnPlEwL5PVOopHwdRSUMdUfYkIHSh1AJ409/sqcB1Nsw\npKEzRqr2DUJcSUv2OZ+XFJGZgHv1ylaSA/mM3kJkq6onTKgvs4qLLgJmzqxPy2bDsOWvq6Rc4Kik\nQs7g0fHaa7XpT5kyCTNnjgUwFhMnjo1P2ADvrj1CiJ1sz4hoDhENEULMIaKhAN4xBJsJYKTye0Tl\nHoQQc5X7VwL4u7s0Y/Gd7wDXXJNJFbr6JytT/XXITDEFYZi8pELg2lvf9i6ywcm8VfI0lV3f+C7E\nS0on5ZSEwZUwirJhyJmirv6LUUnpEsagQWHx1TSAbOCRW2B3d5v3ouIOuPL/vfdWz7K25evrP7qE\nEaKS0k+qyyNhbLpp9gfY0+nXD1hjDXv6PgnDppJywbXVj54+Byb7kHrvi1/sxPDhnZg1K1tz9c9/\njuMn7kFeldTtAI6oXB8O4DZDmCkA1iOitYhoOQAHVeKhQjIS+wF4PiRzSRimQVI+B3g6RAmbSkre\nX201Xjp6elzVCmBvYHffbZ+9SYnB5G4ZM9i5numEkVIlBaSRMDhlcUkY+gCXQsJwnc1hy0NXSalt\n3nSELHeCos+QXYSxyir+Mi5dWjVq2yRu267KEpxZdt51GC4JBeBvfCnL8dBD/rCc8cf17qEG/VY2\nep8HYCciegXADgB+CQBENIyI7gAAIcRSAMcBuAfACwCuF0JUtJb4VUW99TSAbQGcxM1YNkoXYZgM\nwRK2gd4lYcycWW/cMh16r177DMIhKqlnn7U3BkkYJttF0YThykdddMQtQ69e1XMcfGH32KMahyNh\nPPFENb7LhgHk73j6YOxzd7RNBiRMu9WGDia2mbLrG6v6eVsZu7ur27XYysAhjFiVFDecL76PMEzf\n0geTSkpHHglDv1ekW23kUTUZhBDzAOxouD8LwB7K7wkARhvCHRaeZ/Xa5A2gVpycgYUQhk3CIDKL\nsnlVUiEShis9nTB8KiluuqZnIRJG7Mx8yJDqtZrGiBHAjBn1KoFQLympkurVy+wlFSJhrLde7bG1\nepyYQUbCJGHI8vm8ZVzwEYZM+4ILeGVcurRqo1EnERI2lVQqCYPrpuqrH64NQ/6/6ip3eKB4lZR+\nTx0XW20dRlMhZzY2CUMnDM7eUaE2DN9z36BtkjBcjdZn9G41CSNkkFTTsaUpVQC6lwqXMNT0XOsw\n1LRN5VCx7bbm+y4Jw1c20z3bAUq+NFxls5G9fPdTTrGnf8MNwNZbV/uhXGfCld7V/IH8Rm9OuLwS\nhkSo0dtXFtfAHjroF6mSyiVhNBOSHFyLmCRhyEbAqUg9jL7OQgfXrTZEJOWkp0J2WDWdUAkjJE8X\nYcQsoNLD2tR7pjKphPHee2F5+PaS4hq9bQvSQiUMH+HqW4Pkkeh8ZeMMuN/4RjXu0qXVfjZ/vjl+\nXqP3iivyJYxQwuDaMPT64hi9OXtJhZACZx1GTLqsvNMm11joA5QuYUi3Q/lRYwjDdl/q2ItQSYWo\nhySKtGE0SiVlkjBsYUyEEZqHby8pdSt9F1ZdtTZtCb1eYnYu1t1b1UlSCpWUWsaNNzbn60ufqHbR\n3fz59WE4KinfEcHz54cb9U1ldSHUS6ooCWPCBPsz3zu0stG7qdAlDBthhHjxmHSvpvvDh2e7uPrW\nYcQYvUPUQxJ5vaRSudXaBk0fbGooX73J+pVrcTj5yP+LF9slDKBKBD7YwrkGF9NZLD7ClXU9bRqw\nYEE+gtb7BBEwZUr9c25akjDuuSfb5p4rYYTaMIpWSXEnHiEShs3ord7T351zVo8tLZNtNxXaljDk\nAMNRSenE4UtXhU0lpS5Y0p9vvnl9enp812AeqpJSy+myZbgQQlI2wjA1VO7Ao6vQfPWmd9jNNjOv\nlAaAN9803//008wDyCZhmPLVvb6GDKl3tdbjSMgy33BD/a7IpvBqHHlNBJx4Yv0zVxqustnIPlTC\n6OrKvLZ22sk+SzelyVWF+crVKBuG3v64Eoavn8cavW+6qf5eqZJSoA+ELgljdMUva+21s/95VFIu\nwtCfyT1+XOn5PKEOs/iPtaJbbZEqKX0g099JdnCi7HAt9VwAPc4551TVLkRVwlCN3r7vv/32tfef\nfz7bfvqYY/wqqRhVgZqmPHtCT98WnpOuOvCFSHYqZBqudSY2FVooYRQlYfhsGJyFezZXcFuaalkW\nLTKHUfM2xTvggPrwpUrKAkkYEiphLFhQPTRIXVDESVOFywtKJYxVVgF+//v6sLbBVD+PWR/oTYuy\n1PT0cjTTrTYkHxtCVVK6hEGUHRR16aX2sHvtVfs9Fi3iSxh6WhKrrZappP7wB/M7qf9DVKOm/Do6\n7OdguO650g2VMFxp+dZrmNLUzznn5hUbzhffNrh/9rPmdFTCuNCysRHHhmGy+9jg++69epUqKSNc\nEkZHB/DFLwJPPplvpi07qO4Jo3ppyfx+8IP69GwN1LXvTkdHVSripteqbrWx8M0Q9Rmei9RsHezT\nT7PvYLNhuPLlwCZh5CEM1TCc2ujNSSdWwrDF9Z2hLbH//rV56UglYdhsEscck6ndUqqkVCxY4E9H\nwkcYpUrKAqkztw2SHR2ZPSHk45rCPPlkvT+6SyVlSk8P49uo7eSTgXnzeOUDqhu4tZqXVAhCJAy9\nbDbJ0JYuEGfDiFWxAXHbwNvI2ZZOKGGoAyBHwnAN+iphmL4HR8Kwlf/mm+35c8vJie9SH6nusSaV\nlC1tjoThIgy9Leq70+ooVVIK9IFAFdFtahh9VuCCqaFvvnm9hCEXTvn28I9RSXV0ZA3RtFmd7R32\n2MNNnj646qZREoZJPWLLO2TW7pMwpPpPX4fhypcDk9onFLqEYXvmumeCqf50tQYXJsJw5amCu1BO\nIlTCCHVJ5ZbHNKbY0ubYMEIkgc99LvMKtG1jXhKGBT6VlEQeCcNkVFLz50oYOlxG79D01PAmHXcM\nUXLD+ggjz4zcFT9EpWJLV9owNtkE+OMfM5WDr65COqOtXmI9YnQvtJj6VlWorjKGzNBNZ0K4VFem\nuFxwCSNEQlKRZ+GeLW3bO6rhR440h7HhiSdqvf9sWyYt8yopWcmDBtUThvrcN2D40pfYYANzON2G\nYYNtBuezYfjSs6GIld4+CaMIlZRvwNI7LkdC0tuElDCIqlJGHpWU/sxWLyFnH6jvxRlcY1VSep3n\nVUnpiFFJffJJ7YFHtvxd5eQSoc+GYUuXOxlzkfuttwL33WePbxpjVl+9dm87m4TBXZ/ERdsRhhDA\n++/XnofhGySLmGkLAbzwQnZaGFcl5SKM0IV7e+9tft4uRu+TT7bnw5EwJGx2AU59ShsG4CeemHdM\nIWG4VFKuPLnhbG61qW0Ytnsuwlh+eTsJ62i0SipEwrBBhu/b1+4VCfDai40wQg6P46DtCAOo7oqp\nz/BthBGikuIiZu8nLmG44JvNF2HDMOnOTz21eq2WifsehxxSf88027WRlfzPMSSb6kyqpPRDf3zf\no9GE4VJJ+cK7kNetVhqh1bB5jd4cqOkefzwvnPo7tUqKY8PwpaXbI2JgG4/ySv062pIwJEwqKfWZ\nfp268jjwGb1j01Ohpm3SXRYhYRx0ULU8eWbfpnuuBm9TCbikIJdKyiZhcNdh2PI0/ZaIVUmlnPD4\n6s8nYaiqDq4NwyWdADwiVdPYSzkk2qdCnDzZXgYVXMIwaS1C+pCEENl6MVe+oRJGkWg7wvCt9DZB\n7xShsxpOOWywqaRcXlIu+CQM1U+/SJWU2mFss9QQ6FKFr+yumZ4trH5P2jDU+L78XO8n7V0DBrjD\n5lFJnXNOdq2eqx4D18Bn+q3HUyEHfVmXtnAcCcPXfmyTCZ9KitsXUhnhQ+Oq9fCPf4Sn1SjCaNvt\nzYHs43d3A/ffn/32qRFaSSWlEwYXvoFNXcCXSiXlI4wYlZRLTcMhL/lfzmpD4kgsXFg9w8EnYXDq\ncq210s8GdTXRF76QXYdIKSbo78uVjkzP5SCrDnp6e7KtseCuwzDlq+bhs1X4vp9vaxBbOnm0FqY6\nBMx154PeHsqV3gaYKrZRTBs6MOoNS7dhfO1r2dYWPnAJY+21eY06JIz622QsTamS4qT50Uf+2bye\nrn5v2LDsP1fCiHGr1ZHCSypvO/cNfDEShu8s9hi32hg1j+kZt43mUUlx4kmsuGK17en55iGhotHW\nhJFH/dGocvgWRsl7++xT3V7adXKcr+FLldTddzdHJWXDOedkW19z0nTdV/MaMCCO8NR7Q4dm//Ut\nRjhG7x12sOfpQh4JQ2L99ePy1tONJYxQCcN2jztA+5751mGkIgxXGULw2mvAww9Xf7skDA7awoZB\nRIOI6B4ieoWI7iailSzhriaiOUT0bEx8iauvzo6EdJep/l4RlclVobgW7fzmN+YGOnOmPT2bx4+E\nnL327x82mMYac1VpwzUw9+tXO6PSw+rbLvjKHqIacEkYgwdn/0NtGFdfDdx7rz1PV1p53WoXLgT+\n53/scV54IVv5z0nXZgMKGQgbqZKylVGt09//HjjpJHNYX/pFrMMwYciQ2nNUXM4NMSqpopBXwjgd\nwL1CiNEA7gNwhiXcNQBMR4Jw4wMAjjyy1hNDh811rwjkNXoTZecamOJOn25Pz+RGqpYl9AClWJUU\nNw3TGhk97NprA+edF1deW8e1SSU6pCtoChsGF3m9pFZYwd3ON9yw6npug49wfcRpKmOIhCEPCMpj\n9LbZMH7wA2Dddc3xfCSeSiV15JHmeBzYwl59tT2O3l5b1YaxN4BrK9fXAtjHFEgI8TCA92Pjc2Ej\njNDFO9y8OLARhg0XX1xd42CCL41QL6lYwjARpo+sbYQxZkw2wIVIGHo6ejjf3joyvK6KkvdtbcYn\niZji6FDT/va3+WmkmAzpXkMhtgr1vk8l5ZI2gepW4KE2DFsb4qJRKinX4O5Dj1VJARgshJgDAEKI\n2QAGNzh+DXwSRspKDfXq8YWTOO64qieMCandajkDoI8wXGnYzol2SQCuOuNKGK446rUcILiLO1NI\nGqqE8cc/AiNG2MPGrsPgLvALVa2Y3jvGhmGSSkLh8pKS2HHHzN7DlTBCv78vvL61CQcxKqmWcasl\nookAhqi3AAgAPzYEz1tsZ/yxY8f+97qzsxOdnZ11YfLM/EIQo5JKAa4NQ73PIYM8C9Js+SxZkg0m\nhx7qVkmZ0vGpSvIShkSohJFCtRBr9M67GliFz602RFqU5fJtk2GaMKRah2HDaadlf88/X18GX/qu\n59z2MH48L1xIGUxQ29SkSZMwe/YkAIAyZCaBlzCEEDvZnlUM2UOEEHOIaCiAdwLzD4o/1vP2XBtG\nM1RSqfLlutWqYVKTqE0l5SIW/dvklcp8hOFLQ5/h2g5ikrCd7R6DEH1z6pXepgFv6NDssDFTOFt8\nFbqaC3CrAdVr7gFKvrKESlSp4PsmMd8sJs7aa1evOzs7MXRoJ955JyOMcePGhSdoQd4meDuAIyrX\nhwO4zRGWKn+x8b1olNF7m22ydRMSpkb4yiv1z/LqXwH/+6kqKU5enI5kmm3bVFIuyctXFyYJw1Ze\n/bdLjDe9o40obBKGdHVOaUfgIJYwfNKNWidvvw387ne8+Ka69G3NI8PI8o8dW0/Y3P5ga0ONIowQ\nCeP++4HPfz483xiV1ODBmatu0cjb/M8DsBMRvQJgBwC/BAAiGkZEd8hARHQdgEcAjCKiN4no2674\nedAIwnjgAUDVhpkag/TSyEsYemPxrRCPlTD0ettxx3rXRDXdlBKGqxPGShi2ckjohCFdHG1eUpIw\nUkhtIV5Svll7CpgIOmSA4xCGeu/UU812D1M8VzliCDSFmtBXhn79sv+dnWlVmT6ssw6weHF2XZRN\nI9fWIEKIeQB2NNyfBWAP5fc3Q+LHl8ctMkukFktdKo8+fbLBNxV8u7OGEoYcyPUwcuX5b35TH8em\n73ZJBi4bhkvCIDITTYxKSr2n6919xkmdMDjgDk62zn3MMelVUlyELNwzEaDL6E1kJ4wQtIJKypSe\nuiAvBnm+c+ymplw0sAkWD65bbSOgDmQTJ9beywOTCK9ex2w+aBvoXaudbRIGVx8fooJwzVZDJAwX\nicgjeN+vOH/r7yHrtQgJw1ZnF1/cGAnDBK5KEOCrpNS60wnbVq+77Va7WNdWH61kw0il9pIw2Yia\nhbbefNCERhm9uemFbjPuQ8imhSGEEbLSl6uSUuFaxesjBBfRxKqkbGWZP998X7dh5PmW+uBmU1Hp\n7x7iJZVqrynOc1NePglDTnx8Esall8aXwRcnBUImPlw02k4Wgh4nYTSDhRtJGBwjoT6oxUgYIbpj\nziDK8ZKyxXfN4mIJwwZJDJy9pHzgqqRcNo1mqaR85K6+QwxhcG0YrnLFfItG2DBicfnl5jSboSGx\noccRRiM7lYSr4abe44VDGD4xX4UkWRNhxKikXPDZMGyqjRgbhgm+1d+A2ctML09ecG0Y+ndJ4SUV\nG84FDmFwVVI+2NpQs1RSrqNVQ7Haatn/VlA92bBMEIa6nxFg/yATJsTly1UVpJQwXOmmkDAA90Am\nd3nl5sMxeqvkahvc8xCGCltZGyFh6JMIrkqqiMmQ7T3zrio3vbtLwuC21XXWAXbemRfWVB6OhPHS\nS7w0P/64dkNNPS/fPVuYZkx6uWjhosUhz6C8i2l7RAa425HEitMq9HM0XOXhdsKOjnAbxsiRZp9+\nF0IIQ73Oq5IKgZQwfANpShuGi5xSL/wsQtfP9ZIySRihhNG3L3D++fV5pJQwRo/2hwHCpIsvfAEY\nNMgdxtaWY1RSpQ2DiZiB44QTgGnT4vN0SRg2wth//7i85O6qqVRS8nmemTMnLse12UYYLqKNVUn5\nJAwbUgzaXAlDz6+rK3/eOmyDnu89fTYMPb4q/btsGCEoSiVVhEpozBhg3jx3GNtkpLRhFITY3WrX\nWKN+O2QOHngg+8+VMFR861v8fAYMyPbPB6runy6Ebrdgs2FIpFjH4krDZ0gtQsKwvYNPwkihkoq1\nYXzyCT9vTpmEqB5R6wqXN58VVwQOO8xMGHpbVc8F96WdYuHed7+b/f/61/lpFYWi7CwpsUwQho5U\nH0TmFSNhhGDuXOCqq7JrKWGoSGHDMKmkUgyaJsybF6+SSmXDsIErYeSpgw03rP3tmkHGEkbRbrUq\nPvvZ+nvye+y6a7ZNzrXX8ozeW23FL1sKCeOKK7K43zQuLY5DbNuQ8XSpq5UkjB61DqPRbrX6jp8m\npCCMvn2r6y+khCHzPO64+g4bQhguLykThg/nb1Rnw6BBwIIF9fFtRm+XEdFGGKY4ISopWyeVKpxY\n1YYpXa4nXQoJo4jB56tftUuQd91Vf09tb6bvtsYa2f5WPqQ0ercCZNlS7kqcGj2OMEyD97bbAr/6\nVfV3qkbD+cCp1mHIPHQbxsUX28PmtWGYBpcZM3jhfGF8+fkkDP03x1DI+RZLlwK33JKpUUxQj9X0\ngfu9uTaMEMLI28bzSmzcFdAx5YyRMGLyeyd07+2csGksZFtsBbLrUSopwNxQ+/d3n4GcN68YG0Yo\ndAnDhaK9pGLAsWE00q3WhiVLgP32s3vMyY3lOGgmYcTi8MPr841BKpuSK2zRW4O4joMuAqYJ6Nln\nA3vvHZ5W6SXFgE8llVon3yiVFFAlDI6XVKqV3jGztjwzPY6XlH4vhDC4EoYJV1zhjpcHnDrr3x/Y\nYov8efnK/6c/8cL5yhwqYeSRNDjliZlYxA66KW0Yo0a1hmQh0aNUUkBjjd6NVElJ7xEpYdjSevDB\nqudLETaMPBg0qLrgr2gJg0OSJpiM3ssvD2y3HbDeevz0uWEAntH74495aeXJy5RvLLjxYwgjZvBv\npA0jL9Go48mAAfnLkxI9ijBitwZpJwnDRxhf/Wp9frE2DC5McXWVjhDZwDtrVm2cyy+vrtwNkTAk\nON9ALYMvTdsBVOutB0yd6s8jBqm3j0mBIlRSqSdqMRJGK83WdeiEMW1atrK9ldDjCKORDSLUqyGF\nhGFyq80LovQ2DH2bFdd6FGkX4KzDSGXDCJEwuFtfxISxpR+aRmr4bDXcATrmHhc9jTB0o3fM7tzy\nlgAADhhJREFU2jCJ0obBQLPWYTTD6J3qHYYPz2/DiIGpA6faGmTFFYFNN62Pw3kf3+aDRYBr9G4U\nZs/Otn7Jg0YYvVUUQRgx2opf/ALYYIPweED4BLQZ6FESBtDYjbs4C/e4qhUfJGHI/1/7GjB9enx6\nEk88kS0kcxFGzGxSh81Lyubp4tuDy0UY771n3nJCTfOQQ8xeRyuvXH8v9cy41SF3FciDIiWMPHG4\ncadNi7MfnH56eBwJk9G71ZBreCWiQUR0DxG9QkR3E9FKlnBXE9EcInpWu382Ec0goicrf7vmKU+j\nbRih6pAUKilJTptuCtx5Z1y+srwrrpitKzC51aaGjTCKkDD69vXP0tZcExg3rvbeW28B99/vjheC\nFKRi27qj1VEkocbYMPS4PuRRB8WiHSSMvMPE6QDuFUKMBnAfgDMs4a4BYNsL9kIhxBaVv8gNxjMI\nAey7L3Drre5wqVVSjfCSkpKFzCvPRnR//nPt72aqpDgSRt4T9yR8M8YRI4DVV/enw0XImg1b+fv1\na62tISRawa02BO1gw/ARRiuUPS9h7A3g2sr1tQD2MQUSQjwM4H1LGkmq4StfAU47LZtd2ha6yEa+\n2mpmNUcoGmnD0CWMlYyynBnq+/Xvn6lj1LLldavNo5LiSBihKikTFizICCEGsUbvrbbK9lHiILWE\nN3x42vRC0YiV3ip8C1pTEoaUzFMj1oGjkcirLRsshJgDAEKI2UQ0OCKN44joUACPAzhFCPFhTEF+\n/vNal1IXHn4YWLw4JpdacETI1DaMXr3SzziL8JLi5Knnw1mHYYvv62Qhs/1UIMoWXnHDpsRPfwqc\nckr9/UZJK400ej/3nF+FlJIwbOe+54Vs/ynK2DQvKSKaSETPKn/PVf7vZQgeWsxLAawjhNgMwGwA\nFwbG/y9CvFsGD3bPNm+/nZdOyBoAIK1KKiVMEgaQttGFShjN2hrEBFPdpF5QlZow+vRJq14LRSPc\namUb2XjjxkoYRcG3HqcVjOHeIgghdrI9qxiyhwgh5hDRUABB23UJIeYqP68E8HdX+LFjx/73urOz\nE52dnTjhBOCii9K6Q66xBi8cR8JIvflgEQ0+rw0jRiVlihuzcK8ZYvzUqemNoo0qf6MGzCIljBi0\nO2E8+iiw5Za8dCZNmoS5cycBAJQhMwnyctbtAI4AcB6AwwHc5ghL0OwVRDRUCDG78nM/AM+7Mhtr\nePvf/S4jjJSrZbmNqpE2DFmmIlYFd3Tks2Go4L6vbzdZk5fU2ms3R8LQoW4RkgqtNpD5vmOKhXv7\n7gtsv709vC/vkL7V7oSx9db8dDo7O7H66p14772MMMbp7oA5kLebnQdgJyJ6BcAOAH4JAEQ0jIju\nkIGI6DoAjwAYRURvEtG3K49+VVFvPQ1gWwAnxRaEI2Fwj6O0bW2tI5QwZOfIg5SSlGr01t9h5ZX5\n2xKk6oQ+t9rXX6/Psx0MhRw0qvytZMMYPz6bBOj45z+ByZP98ZclwghFUd85l4QhhJgHYEfD/VkA\n9lB+G8+zEkIclid/FZyB9Etfqh10bBg9OvN//+gjd7hQldT3vgdcf331aNdQDB6ceXilhq6Smj49\nW6cg1VK+tQkx52H4JIw8C/faEa08kJng++Y77pi1dxXcd+SulC4Jo/Fo826W4aSTMjLwgcg8ozE1\nIo5bYoxKKk+DnTOnmN0rdcIYOTJsURTHEM/p3CEL9ySku3G7E4YsvzyDYZNNmleWFBg6FLjsstp7\nHFsUBzEqqZh8Go12IIwWsLvnx4XRvlX5ELpwT43TSPg6iUklFYJNNwXuvTcsju+oUq6EoR9Z266Q\n737rrcCbbwJHH93c8vQ0LGsSRkuqpHoyOBXOUYeklDCKgs2tNiT+Dju4w+j1MGhQ7bG5QK3boO+I\nVp0wWnk7BQ7kyv1dc22O09pIJWHIsDEqqVaeWLSDhNHC1df6aBcJwwZZNpOXVFF5SXR01B+bq/rS\nq+FtzgpAdbv3VqrXGOTZ6mVZg2wbI0dmR5hysKxJGEWhzbtZGsQ2op4kYbTCgKsShtp5LrwwW83r\nitMK5Y8FUXsMFnnhmljF9Is+ffjrDErCSIM27mbNR4yY24oNNq9KigOO+kA9HEoNP3BgtppXRU+y\nYchV/O2EGB25a6VyuXCvJIy2Rox+lJNOKw1ssmyf/SywyirF5fOjHwFHHOEPZ5MwOHFaqV5D0YqE\nUYTRNJWdKY93VEkY+VAavVHsqma9EbRSg5XlHz++2HzOO48XzmbDMKGUMNoPzdwLqRW3iNfxta8B\nl1ySJq3yiNYGI2YxGidMOw9sRSNGwpDrMNphQLBBvkM7Iaa+U9swehoGDACOPbbZpXCjlDByoKct\nHPLh5z/3h8kzcO++O/DGG2HpyPoscpZe9DdrpITRTGJthd1WS+RDOd/NAd+WykB7eEn5IN/hDNt5\niolw8snZWcpAmD5XiGLPuyh6kF1WBlLXe26xBfCZzzSuLCXisIw0VTdMg/gBBwBTprjjrbwysGiR\nO8yyppJKRYjtrGIKRTuqpGLgUkndcUexRt9lxU4kUa70bjDOPZcXTu79Y0MrGL1Ned52WzGdKFVD\nbQePkVQYOTI7Y6OV4DvONrVbbdGkudxyy9YkpCiUhIHGDuKtopLay3ReYgthWerc48dnZ463Eo48\nMpOyU6KZbrUl0qCHK0iaj55gw2gGfBKGT7JLiaK/2cCBwLBhxeYRCiJgpZXSprms2GpaAaNGFWPX\nW+Y/4W9/Cxx0UHHpl4QRBxdhNHqGWc5o6xGz0DOVhFH2IT9uuQVYvDh9uss8Yfzwh8Wmv6wZvVNh\nWbJhtBs++aR2GxcuUkkYJYH7sfzycd/Ih3L4Khh64/7855tTjnZDKw0K5Yy2FrEDUSoJY/DgNOmU\nCEdJGAVDH/j+53/ab/bs85hRse66aaSoPfcExozJn06JWpx5JnDDDc3JO5WEMWhQa00oliXk6tpE\nNIiI7iGiV4jobiKqM5MR0Qgiuo+IXiCi54johJD47Q59dtqInWFtWHvtbKPBUIweze+gN9wAvPde\neB46Dj8ceOyx/OmUqMWIEcA3vtGcvNv9kKsS+SWM0wHcK4QYDeA+AKa1wEsAnCyE2AjAFwEcS0Qb\nBMRvW/zrX8D556dPd9KkSVHxnnsOiIzKxgorZAsaG4XYuuiJaPW6aKSXVKvXRbsiL2HsDeDayvW1\nAPbRAwghZgshnq5czwfwEoDh3PjtjC99CVhttfTpxnaG/v2L3UKjGSh6YHj0UeDBBwvNIhlafZBs\npITR6nXRrsjL+YOFEHOAjBiIyGmOIqK1AWwG4N8x8UvEoTTaxmPrrZtdgp4D/RCsEu0HL2EQ0UQA\nQ9RbAASAHxuCWzXdRNQfwM0AfiiEsK1rLU1ZiXHMMdnGbiVKNBuf/WxprG53kMjxBYnoJQCdQog5\nRDQUwP1CiDqzKhH1BnAHgLuEEL8LjV8JWza1EiVKlIiAECKJniGvSup2AEcAOA/A4QBus4T7I4AX\nVbIIjJ/shUuUKFGiRBz+f3tn92JVFYbx3xM59KWWF2k5OikS0k0xlUVTBAUqKV6GEAX+CV6YjX9B\nN1FddNMHRvZh3zBBREgXkRd9oDJilickGyeaiGyiLvqQp4u1Tp6ZnGmfpuPZ7PX+4MBe67DZ633O\nZr9nrfU+5yx0hrEMeA1YBZwC7rf9k6RrgGdsb5U0AnwIHCUtORnYY/u9uc5fUERBEARBT1hQwgiC\nIAjKofZOb0mbJX0h6YSk3f0eT6+Zy+g4n8lR0qiklqTjkjb2b/S9QdJFkg5JGsvtIrWQtFTS6zm2\nY5JuK1iL0azBuKSXJA2UooWk5yRNSRrv6Os6dknDWb8Tkp6odHHbtX2REtpXwBCwCDgCrO/3uHoc\n8wrgpnx8BfAlsJ60z/Nw7t8NPJqPbwAOk/ajrst6qd9x/M+a7AReBMZyu0gtgOeBHfn4YmBpiVrk\n58FJYCC3XyXtgRahBXAnyZ4w3tHXdezAx8Ct+fhdYNO/XbvuM4wNQMv2Kdt/APtJZr/G4vMbHQeZ\n2+S4Ddhv+0/bXwMtkm6NQNIgcB/wbEd3cVpIWgLcZXsvQI5xmgK1AH4GfgcuzxWYlwKTFKKF7Y+A\nM7O6u4o9V6Uutt3+I+oXqGCcrnvCWAlMdLRPc84l3nhmGR2Xu8PkCLRNjrM1mqRZGj0O7GKmR6dE\nLdYAP0jam5fnnpZ0GQVqYfsM8BjwDSmuadsHKFCLDq7uMvaVpOdpm0rP1ronjGKZZXT8hX+aGhtf\nrSBpCzCVZ1zzlVU3XgvSksIw8JTtYeBX0m+xlXhfrCUtUw4B15JmGg9QoBbz0JPY654wJoHVHe3B\n3Ndo8jT7DWCf7bY3ZUrS8vz+CuD73D9JKktu0ySNRoBtkk4CrwD3SNoHfFegFqeBCduf5fabpARS\n4n1xC3DQ9o+2zwJvA3dQphZtuo39P2lS94TxKbBO0pCkAWA7yezXdM5ndGybHGGmyXEM2J6rRNYA\n64BPLtRAe4ntPbZX215L+uw/sP0g8A7laTEFTEi6PnfdCxyjwPuCVAhyu6RLJImkxeeUpYWYOevu\nKva8bDUtaUPW8CHmMU7/Tb93/CtUBGwm3SAt4JF+j+cCxDsCnCVVhB0GDmUNlgEHshbvA1d2nDNK\nqn44Dmzsdww90uVuzlVJFakFcCPpS9QR4C1SlVSpWuwiJcxx0ibvolK0AF4GvgV+I+3j7ACu6jZ2\n4GaSoboFPFnl2mHcC4IgCCpR9yWpIAiCoCZEwgiCIAgqEQkjCIIgqEQkjCAIgqASkTCCIAiCSkTC\nCIIgCCoRCSMIgiCoRCSMIAiCoBJ/AWv5pBwdRVYsAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x106c92940>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXwAAAEACAYAAACwB81wAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nKZKj47Fk1noAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1200b09b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "initial = tf.truncated_normal([1000], stddev=0.1)\n",
    "with tf.Session() as sess:\n",
    "    result = initial.eval()\n",
    "    # print(result)\n",
    "    plt.plot(result)\n",
    "    plt.title('Showing the raw initialized data')\n",
    "    plt.show()\n",
    "    \n",
    "    plt.hist(result, np.linspace(-0.2, 0.2, 11), \n",
    "             ec='none', rwidth=0.9, color='#47c0fc')\n",
    "    plt.xlim(-0.3, 0.3)\n",
    "    plt.ylim(0, 200)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-0.2 , -0.16, -0.12, -0.08, -0.04,  0.  ,  0.04,  0.08,  0.12,  0.16])"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    ""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    ""
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [python3]",
   "language": "python",
   "name": "Python [python3]"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3.0
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}